Wi-Chat: Large Language Model Powered Wi-Fi Sensing

2025-02-18

AI Summary

This work addresses the limitations of hand-crafted feature extraction and conventional signal processing in Wi-Fi sensing. We propose, for the first time, the direct application of large language models (LLMs) to raw channel state information (CSI) for end-to-end human activity recognition. Methodologically, we design a physics-informed prompt engineering framework that encodes Wi-Fi propagation principles as structured knowledge into the LLM, enabling zero-shot inference without time-frequency transformations or manual feature engineering. Our key contributions are: (1) establishing a novel paradigm for integrating LLMs with wireless physical-layer sensing; (2) empirically demonstrating LLMs’ implicit modeling capacity and causal reasoning ability on non-linguistic physical signals; and (3) achieving high-accuracy zero-shot activity recognition on real-world Wi-Fi datasets—outperforming all baseline methods significantly in accuracy.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Recent advancements in Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks. However, their potential to integrate physical model knowledge for real-world signal interpretation remains largely unexplored. In this work, we introduce Wi-Chat, the first LLM-powered Wi-Fi-based human activity recognition system. We demonstrate that LLMs can process raw Wi-Fi signals and infer human activities by incorporating Wi-Fi sensing principles into prompts. Our approach leverages physical model insights to guide LLMs in interpreting Channel State Information (CSI) data without traditional signal processing techniques. Through experiments on real-world Wi-Fi datasets, we show that LLMs exhibit strong reasoning capabilities, achieving zero-shot activity recognition. These findings highlight a new paradigm for Wi-Fi sensing, expanding LLM applications beyond conventional language tasks and enhancing the accessibility of wireless sensing for real-world deployments.

Problem

• Integrates LLMs for Wi-Fi signal interpretation
• Enables human activity recognition using CSI data
• Expands LLM applications beyond language tasks

Innovation

• LLM-powered Wi-Fi sensing
• Zero-shot activity recognition
• CSI data interpretation